Agreed on both directions: my position is that actually doing the analysis rather than going by gut instincts is less common than it should be for a field with pretenses of being a type of engineering. Cost analysis is especially big on that since most JS hype tends to be “3ns faster on this benchmark!” or “look at how easy this trivial task was!” rather than, say, development velocity on overhead on a non-trivial project over a year+ timeframe.
I agree with the lack of analysis portion being a common thing, but that's something you have to work out for yourself as a part of the process. It's not as if React is the only faddish tool over the years - I'm sure there are plenty of older projects that are only using jQuery because it was the new hotness. And even more that were initially done with jQuery, but would probably work better with React if it weren't for the Lead's disinclination to learn a new framework. There's plenty of fad chasing, but there's also plenty of bull-headed resistance to new tech.
The question of if React or jQuery or Vue or whatever is the best framework to use for a specific company or product is tangential to the question of if someone with only jQuery experience can be effective on a team with React as its core tech. The decisions have been made at that point.